PROJECT SUMMARY/ABSTRACT Diabetes mellitus (DM) is a strong risk factor for cancer development. However, it is unclear if DM is also a risk factor for transforming pre-leukemic stem cells (pre-LSCs) into full-blown leukemia such as acute myeloid leukemia (AML), chronic myelomonocytic leukemia (CMML) or severe form of myeloproliferative neoplasm (MPN). To this end, extensive whole exome sequencing on peripheral blood (PB) cells of more than 20 thousand persons who were unselected for hematologic phenotypes were examined for 160 genes that are recurrently mutated in leukemia. Two most mutated genes found were epigenetic regulators: Ten eleven translocation methylcytosine dioxygenase 2 (TET2) and DNA cytosine-methyltransferase 3A (DNMT3A). Notably, both these genes are mutated at very high frequency in patients with MPNs and AML. Importantly, epidemiological findings demonstrate that persons with DM are more likely to have these mutations than those without DM. However, there is a significant gap in knowledge regarding our understanding of DMs causative role in MPN and leukemogenesis. Studies using murine models of loss of Tet2 function (Tet2-/-) show that Tet2-deficient hematopoietic stem cells (HSCs) have abnormal global 5-hydroxymethylcytosine (5-hmC) levels and local 5- hmC levels in genes responsible for self-renewal of HSCs. In addition, Tet2-/- HSCs show a competitive advantage (i.e. clonal hematopoiesis) over normal HSCs but do not progress to AML development. Moreover, mice with haplo-insufficiency of Tet2 (Tet2+/-) manifest milder form of these phenotypes only when aged, indicating that Tet2 functions as a putative tumor suppressor. We discovered that Tet2-deficient mice manifest a sustained response to acute inflammation and aged nave Tet2-/- mice show elevated expression of a series of genes encoding the innate immune/pro-inflammatory pathway components including S100A8/9, TLR4, NF?B1 and IL-6. As patients with DM develop chronic inflammation and have increased expression of S100A8/A9, the central hypothesis of this proposal is that diabetic individuals that carry a single genetic lesion in the form of TET2 in their HSCs (in absence of any hematologic malignancies), will be more susceptible to developing severe MPN and/or myeloid leukemia, in part to an overactive pro-inflammatory cytokine signaling cascade or a forward feeding positive signaling loop and in part to hyperglycemia (HG) induced further global reduction in 5-hmC levels in pre-LSCs bearing loss of Tet2, thus mimicking pre-LSCs lacking all forms of TET. Indeed, by introducing Tet2 mutation into a murine model of diabetes (Ins2Akita/+), our preliminary data demonstrate that the diabetic mice haploinsufficient for Tet2 (Ins2Akita/+;Tet2+/-) progressively develop lethal AML/severe MPN. Importantly, treatment of Tet2-/- mice exposed to an acute inflammatory challenge with an anti-inflammatory molecule, APX3330, reverses this phenotype. Based on these preliminary results, we hypothesize that progressive DM acts as a significant risk factor for transforming pre-LSCs and/or clonal hematopoiesis into heme malignancies in an age-dependent manner.